Gulbransen(Seeburg) Select-a-Rhythm white noise

[Enzo]

paused, as in stopped, not going. This is a drum machine basically, and it needs to be able to stop and start. It probably doesn;t call anything "pause" that is just my term. Since your hiss leak occurs even when a rhythm is not happening, we want to troubleshoot it that way. So if there is a start and stop, press the stop.


If you can;t get a good read on teh gate base, what does the gate collector do in each?

At the trigger, does grounding the base at least control the collector on the trigger?

[pontiacpete]

at the gate collector the voltages move slightly 13.95 to 13.55v.

Grounding the trigger gate raises the voltage on the trigger collector 0 to 13-14v.

[Enzo]

OK, so the triggers work.

remember on the gates we are going to see the voltage limited by the "diode" of the base-emitter junction. The emitter and base are already biased up to a 2 junction drop, so when the trigger turns on, the gate base will only rise to the third diode drop.

The whole idea here is to be systematic. Your triggers work. The gates still seem to be hanging, at least in my view. So we need to isolate the opening through which our noise flows. Grounding the base of each gate will turn it off, or should. That should tell you which one or ones is leaking. if it is more than one, it is more than one. Or the other way, the three gates have mixing resistors, I see 100k, 220k, 330k. You could disconnect one end of each of those one at a time to see which one or ones are carrying the noise. And that would tell you which gate was involved. Or gates.

[pontiacpete]

So.. grounding the base of the brush gate takes out a lot of the hiss, grounding the top gate, (is that the snare gate?) takes out the rest of the hiss. Grounding the middle(cymbal) gate does nothing to the hiss.
Same results when lifting the 100k and the 330k mixing resistors. Lifting the 220K doesn't do anything.
Since the brush gate is new what does that leave? The inductor? The mylar cap or the resistors?

[Enzo]

The gate collectors can't turn themselves on. And the drive for the #1 and #3 come from different places, about the only thing I see in common is that cathode reference. Just idle thinking now, but if that 8uf cap is leaky or one of those two diodes is failing, then the cathodes might drop a bit a start to bias the transistors on. I'm sorry, emitters, not cathodes.

And we already determined that grounding the A for sandblock input has no effect?

[pontiacpete]

Grounding the sandblock does nothing.

grounding the 8uf makes more hiss, if that means anything.
I tried swapping that cap no effect.
What are these diodes? all they seem to have on them is a white stripe and black body and a green "T" which appears to be written by hand.

[Enzo]

Well yes, grounding that cap would increase the hiss. That was the point of my theory. The cap keeps the voltage steady, the two diodes establish a two-junction drop potential as bias. I was surmising that if one of the diodes was breaking down or the cap was too leaky, then that voltage would drop and bias the transistors on. By grounding the cap that is the ultimate in voltage dropping, all the way to zero. And then the gates pour open.

I don't know what the diodes were, today they would all be 1N4148, which I see no reason would not work here as well.

Is there the exact same voltage across each of those series diodes?

[pontiacpete]

The diode closest to ground is .700v and the other is .678v

[Enzo]

CLose enough, I'd hope.


I'm not sure what to tell you. You have gates open, either from the parts being leaky or from them being turned on, at least partly. By systematically forcing them closed we should be able to determine just why they are open. At least that is the idea behind troubleshooting.

[pontiacpete]

I think I found the problem. I start by changing out some of the diodes around the base of the brush gate, no results.
Then I changed out the transistor in the noise generator just to the left of the nose diode with the same 2n5089.
Still hissed.
Next thing I try is changing the nose diode. This one reads on it 4000. I have no idea what this is but I figure from the schematic that it's a NPN transistor, emitter pointing away from the base.
So I put it another 2n5089 in it's place and sh works like a charm. No hisss
Is there any reason why I shouldn' t use the 2n5089?

[g1]

First off, nose is a typo for noise .
If you are talking about the transistor labeled "noise generator" (used as diode, top right of page 72), if you made it stop hissing, it is probably no longer working as a noise generator. If this is the case, you will no longer have a proper sounding brush, cymbal, or snare drum. Are they working?
More info about the noise diode and it's zener action on page 68.

[pontiacpete]

That's funny, I thought Enzo was using some special terminology for that transistor

There's definitely something working on those gates. It sounds pretty good.
I guess the question is, would it be better with an exact replacement?
I'm not even sure if says what I said it has on it "4000" . The '4' is mostly covered by a dab of paint.

[g1]

My point was that the noise generator (Q2501) is supposed to make noise and hiss. If the gates are working right, then a noisier Q2501 should not matter. I'm wondering if there is a coincidence and something else got fixed when you changed it to the 2N5089.
Maybe Enzo will have an explanation.
Otherwise, if you are getting proper sounds from snare, brush, and cymbal, I guess it's fixed.

[Enzo]

So you cured the hiss by changing the hiss generator?


OK, so now do ALL the sounds come through? All the snare hits and cymbals?

[pontiacpete]

Yes changing the hiss genny seems to have fixed it. Here's a clip of the what it sounds like now.
seeburg SAR - YouTube

here's another one to compare it with:
Vintage 60's SEEBURG Select-A-Rhythm Analog Drum Machine Beat Box - YouTube